A NONLINEAR MODEL FOR ESTIMATING PROBABILITIES OFkEVENTS1

Abstract
Equations are derived for the estimation of the parameters in a nonlinear model for the probability of more than two mutually exclusive and exhaustive events. The estimated probabilities are between zero and one and sum to one. The equations for least squares and maximum likelihood estimation are given, and it is pointed out that the maximum likelihood estimate has the form of weighted least squares with estimated weights giving more weight to low probability events.

This publication has 0 references indexed in Scilit: